Method for monitoring lithographic apparatus

Abstract

A method of determining a parameter of a lithographic apparatus, wherein the method includes providing first height variation data of a first substrate, providing first performance data of a first substrate, and determining a model based on the first height variation data and the first performance data. The method further includes obtaining second height variation data of a second substrate, inputting the second height variation data to the model, and determining second performance data of the second substrate by running the model. Based on the second performance data, the method determines a parameter of the apparatus.

The invention claimed is: 
     
       1. A method comprising:
 decomposing first height variation data of a first substrate, into at least a substrate specific fingerprint; 
 determining a model based at least partially on the first height variation data, first performance data of the first substrate and the substrate specific fingerprint; 
 obtaining second height variation data of a second substrate; 
 inputting the second height variation data to the model; 
 determining, by running the model using a hardware computer system, second performance data of the second substrate; and 
 determining a parameter of a lithographic apparatus based on the second performance data. 
 
     
     
       2. The method according to  claim 1 , wherein the first substrate and the second substrate are the same substrate. 
     
     
       3. The method according to  claim 1 , wherein the first and second performance data is overlay data and the first and second height variation data is levelling data. 
     
     
       4. The method according to  claim 1 , further comprising:
 obtaining reference data of the second substrate; and 
 inputting the reference data to the model. 
 
     
     
       5. The method according to  claim 4 , wherein obtaining the reference data comprises running the model a first time, and further comprising storing the second performance data as the reference data for inputting during a subsequent run of the model. 
     
     
       6. The method according to  claim 1 , further comprising:
 determining the model based additionally on first alignment data of the first substrate; 
 obtaining second alignment data of the second substrate; and 
 inputting the second alignment data to the model. 
 
     
     
       7. The method according to  claim 6 , further comprising upsampling the resolution of the first or second alignment data, based on the respective first or second height variation data. 
     
     
       8. The method according to  claim 1 , further comprising:
 decomposing the second height variation data into a plurality of subgroups, and inputting at least one subgroup of the plurality of subgroups to the model. 
 
     
     
       9. The method according to  claim 1 , further comprising:
 obtaining second performance data of the second substrate at a sparse resolution; 
 inputting the sparse second performance data to the model; and 
 estimating, using the model with input data, dense second performance data. 
 
     
     
       10. The method according to  claim 1 , wherein the second substrate is a reference substrate or calibration substrate. 
     
     
       11. The method according to  claim 1 , wherein the parameter is a measure of the quality of a substrate support of the lithographic apparatus supporting the second substrate. 
     
     
       12. The method according to  claim 1 , wherein determining the model comprises determining a matrix of weights and biases. 
     
     
       13. The method according to  claim 1 , wherein the model comprises a first model and a second model, wherein the first and second models have different inputs and different outputs. 
     
     
       14. The method according to  claim 13 , wherein the first model is run at a first periodicity to determine a first output relating to a first parameter; and
 the second model is run at a second periodicity to determine a second output relating to a second parameter, 
 wherein the first periodicity is greater than the second periodicity. 
 
     
     
       15. The method according to  claim 14 , wherein running the first model comprises using a calibration substrate as the second substrate, and running the second model comprises using a reference substrate as the second substrate. 
     
     
       16. A method comprising:
 obtaining a model trained to correlate height variation data to data of a performance parameter; 
 obtaining a set of height variation data of a substrate; 
 decomposing the set of height variation data to determine a process signature of the substrate; 
 inputting data related to the process signature to the model; 
 running the model, using a hardware computer system, to determine a predicted deviation of a value of the performance parameter between the substrate and a substrate standard; and 
 calibrating the substrate using the predicted deviation. 
 
     
     
       17. The method according to  claim 16 , wherein the height variation data is levelling data. 
     
     
       18. The method according to  claim 16 , wherein the data of the performance parameter is overlay data. 
     
     
       19. The method according to  claim 16 , wherein the data related to the process signature comprises the process signature. 
     
     
       20. The method according to  claim 16 , wherein the substrate is a reference substrate.